© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 6351
    Prev
    Next
    INDEX
    Explanations

    references to sports teams

    oai_token-act-pair · gpt-3.5-turbo

    references to a specific sports team

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    tain
    -0.82
    pire
    -0.80
    ranging
    -0.79
    tml
    -0.78
     drip
    -0.77
    tar
    -0.69
     Maw
    -0.67
    fect
    -0.67
    sql
    -0.67
    imentary
    -0.65
    POSITIVE LOGITS
     mates
    0.95
    liquid
    0.83
     captain
    0.82
     coached
    0.80
    mates
    0.79
     Antar
    0.74
    ubs
    0.73
     captains
    0.72
     scrimmage
    0.72
     mate
    0.72
    Activations Density 0.053%

    No Known Activations